_ Git - youtube-dl/blob - youtube_dl/extractor/br.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import ExtractorError
   8
   9
  10 class BRIE(InfoExtractor):
  11     IE_DESC = "Bayerischer Rundfunk Mediathek"
  12     _VALID_URL = r"^https?://(?:www\.)?br\.de/mediathek/video/(?:sendungen/)?(?P<id>[a-z0-9\-]+)\.html$"
  13     _BASE_URL = "http://www.br.de"
  14
  15     _TEST = {
  16         "url": "http://www.br.de/mediathek/video/anselm-gruen-114.html",
  17         "md5": "c4f83cf0f023ba5875aba0bf46860df2",
  18         "info_dict": {
  19             "id": "2c8d81c5-6fb7-4a74-88d4-e768e5856532",
  20             "ext": "mp4",
  21             "title": "Feiern und Verzichten",
  22             "description": "Anselm Grün: Feiern und Verzichten",
  23             "uploader": "BR/Birgit Baier",
  24             "upload_date": "20140301"
  25         }
  26     }
  27
  28     def _real_extract(self, url):
  29         mobj = re.match(self._VALID_URL, url)
  30         display_id = mobj.group('id')
  31         page = self._download_webpage(url, display_id)
  32         xml_url = self._search_regex(
  33             r"return BRavFramework\.register\(BRavFramework\('avPlayer_(?:[a-f0-9-]{36})'\)\.setup\({dataURL:'(/mediathek/video/[a-z0-9/~_.-]+)'}\)\);", page, "XMLURL")
  34         xml = self._download_xml(self._BASE_URL + xml_url, None)
  35
  36         videos = [{
  37             "id": xml_video.get("externalId"),
  38             "title": xml_video.find("title").text,
  39             "formats": self._extract_formats(xml_video.find("assets")),
  40             "thumbnails": self._extract_thumbnails(xml_video.find("teaserImage/variants")),
  41             "description": " ".join(xml_video.find("shareTitle").text.splitlines()),
  42             "uploader": xml_video.find("author").text,
  43             "upload_date": "".join(reversed(xml_video.find("broadcastDate").text.split("."))),
  44             "webpage_url": xml_video.find("permalink").text,
  45         } for xml_video in xml.findall("video")]
  46
  47         if len(videos) > 1:
  48             self._downloader.report_warning(
  49                 'found multiple videos; please '
  50                 'report this with the video URL to http://yt-dl.org/bug')
  51         if not videos:
  52             raise ExtractorError('No video entries found')
  53         return videos[0]
  54
  55     def _extract_formats(self, assets):
  56         formats = [{
  57             "url": asset.find("downloadUrl").text,
  58             "ext": asset.find("mediaType").text,
  59             "format_id": asset.get("type"),
  60             "width": int(asset.find("frameWidth").text),
  61             "height": int(asset.find("frameHeight").text),
  62             "tbr": int(asset.find("bitrateVideo").text),
  63             "abr": int(asset.find("bitrateAudio").text),
  64             "vcodec": asset.find("codecVideo").text,
  65             "container": asset.find("mediaType").text,
  66             "filesize": int(asset.find("size").text),
  67         } for asset in assets.findall("asset")
  68             if asset.find("downloadUrl") is not None]
  69
  70         self._sort_formats(formats)
  71         return formats
  72
  73     def _extract_thumbnails(self, variants):
  74         thumbnails = [{
  75             "url": self._BASE_URL + variant.find("url").text,
  76             "width": int(variant.find("width").text),
  77             "height": int(variant.find("height").text),
  78         } for variant in variants.findall("variant")]
  79         thumbnails.sort(key=lambda x: x["width"] * x["height"], reverse=True)
  80         return thumbnails